Hyperthermia, particularly when used in conjunction with ionizing radiation, is becoming a viable modality for cancer treatment. A high percentage of the clinical data currently available were obtained with applicators allowing little control of heating patterns within the treatment volume. However, such control is essential for obtaining therapeutically optimun temperature distributions, particularly when blood flow changes during treatment. This proposal discusses several new phased array concepts for ultrasound hyperthermia applicators which should, based on extensive preliminary studies, provide the needed heating flexibility. Optimun engineering designs will be obtained by computer simulation involving acoustic field intensity computations and temperature distributions in the treatment volume as predicted by appropriate thermal modeling. Prototype applicators based on optimal designs will be fabricated and tested to demonstrate feasibility and to provide the technical basis for eventual construction of new hyperthermia systems. Testing will involve detailed field pattern measurements, animal studies involving normal and tumor tissues, and preliminary clinical trails. Preliminary research has led to several new ultrasound applicator concepts, viz., the concentric-ring and sector-vortex phased arrays. Concentric-ring phased arrays subdivided into sectors (radial slices) can, with appropriate phasing, produce variable diameter annular (or ring) focal regions. Such focal rings can be effective in heating tumors if directed around the tumor periphery. Scanning the focal ring diameter and focal length provides considerable heating flexibility. Moreover, appropriate phasing of the rings and sectors can produce patterns which are circularly asymmetric. By controlling these asymmetries, nonspherical tumors can be heated more optimally. Research on other more advance concepts is also proposed herein.